Ground vs. Neutral

[gordonl]

I have electronic relays with three phase CT's connected in wye after the phase inputs, and run through the ground element of the relay. The system is 3 wire, resistively grounded to 1000A.

Anybody see any advantage/disadvantage to using ground settings on the relay versus neutral settings for ground fault protection? The neutral is calculated based on phase currents, versus CT sumation for ground.

 

[stevenal]

Using the ground input simplifies testing with a single phase test set. Attach to a phase input to test phase settings; attach to ground input to test ground settings. No need to disable and reenable protection elements (with the risk of them not getting reset properly).

 

[bru]

gordonl

I know that there is no differentiation between ground fault sensing and phase imbalance with this kind of (residual ground) connection.

also, stevenal, you can test the relay using a single phase set with either configuration without changing settings. If the current injection is across the Phase CT inputs only, the resulting sensing by the relay will be phase only. If the return current is run thru the ground return, (thru the ground or Neutral element) the relay will sense ground current.

In either case, lower end relays will always have settings changed if the short time delay and instantaneous elements are sufficiently low enough to prevent long time delay verification.

The obvious advantage to you gordonl is that you would not have to install ground sensing elements into the neutral to ground connection and the other is that the residual connection will sense ground fault in the zone directly downstream of you phase sensing, whereas neutral to ground sensing will only work on current coming back thru the NGR.

Also, depending on the kind of relay you are using, the phase imbalance may cause occational trip on energization due to magnetization of CTs. This can usually be compensated with time delay but that depends on system coordination.

Bru

 

[stevenal]

If you apply current to one phase of a relay that calculates neutral from the phase currents, it will calculate a nonzero neutral current. If you apply the single phase current across all three phase inputs in series, you will also get a nonzero calculation. You could apply the current across pairs of inputs in reverse polarity, as long as negative sequence elements are not used.

 

[peterb]

The separate ground element is usually provided so that you can connect to a separate zero sequence type CT & get a more sensitive measurement. This would normally be used for a high resistance grounded system, rather than the medium resistance system that you have.
Another case where it could be used is where the overcurrent relay is protecting the source transformer; in this case a CT on the grounding resistor connection would be used as the input to the 51G element.
That said, I can't see a good reason to use the separate 51G element rather than the calculated 51N element if the same set of CTs is being used.